[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [linrad] cpu load benchmarks

Hi Dmitry,

> Strange, but I found a very different results on my
> notebook(P4M-1300/256Mb RAM/i855/i8x0sound). 
> Avg	CPU [%]
> 100	3.0
> 10	4.5
> 3	8.1
> 2	10.7
> 1	18.5
> With the same parameters as Leif wrote. Waterfall had about 70% of
> screen. 1024x768 (mode 12), AD speed 44KHz. Forced VESA driver.
Your graphics is probably faster than mine. Now your data input is much
slower, with A/D speed=44kHz on a single channel you have only 23%
of the bandwidth as compared to I/Q input sampling at 96 kHz so it
is not quite fair to compare your results directly to the timing I have.
Assuming you select I/Q data at 96 kHz you would get 4.36 times more
load on the graphics which would increase graphics from 15.5% to 67.5%
for a total load of 70.5% at avg=1. This is much better than what I have,
it is similar to what Jeffrey Pawlan, WA6KBL reports for the Sapphire 
Radeon ATI 7000 AGP 32MB. ( I will order one of these - the cost is really
> I810 driver does not work at all. With forced FBDEV driver VGATEST
> working fine in all modes, but when linrad starts svgalib gives an error
> Signal 11: Segmentation fault received.
I have no idea what the problem might be. It seems there are sometimes
interrupt conflicts between the sound system and some graphics drivers.
It is beyond my skils to have any idea what is going on.


Leif  /  SM5BSZ